STUDY OF FUZZY LOGIC IN MEDICAL DATA ANALYTICS … · certain point and long time are vague and...

STUDY OF FUZZY LOGIC IN MEDICAL DATA ANALYTICS

Mrs.Susmita Mishraand Dr. M Prakash

1Assistant Professor ,Department of Computer Science and ngineering

Rajalakshmi Engineering College,Chennai,India

2Professor ,Department of Information and Technology

Karpagam College of Engineering,Coimatore,India

[email protected] ; [email protected]

ABSTRACT

Fuzzy logic provides an intelligent periphery for knowledge representation and reasoning to handle

inaccuracy, uncertainty, and vagueness. Fuzzy systems have been successfully applied in healthcare due

to their ability to infuse human expert knowledge and granular computing, to describe the behavior of

complex systems without requiring a precise mathematical model.This paper provides an outline of basic

fuzzy logic and how this logic can be used to perform various decision-making tasks. It also emphasizes

FL tasks can be applied to different types of medical data, to classify a certain type of disease or diseased

patients, in constructing a decision support system.This paper is a descriptive study of FL and its

applications in healthcare related fields. The main motive of this paper is to draw a brief description of

fuzzy logic applications on various medical diagnosis system.

Key Terms:Fuzzy Clustering Mean (FCM), Fuzzy Inference System (FIS), Fuzzy Logic(FL),Fuzzy Membership

Functions(MFs),Fuzzy Set Theory(FST)

International Journal of Pure and Applied MathematicsVolume 119 No. 12 2018, 16321-16342ISSN: 1314-3395 (on-line version)url: http://www.ijpam.euSpecial Issue ijpam.eu

16321

INTRODUCTION

Medical datum is any single observation of a patient, thus medical data is a combination of different

observations of a patient. Medical data contains information on a person’s state of health and the medical

treatment that they have received.Szolovits(2006) stated that medical data analytics is a term used to

describe the medical analysis activities that can be undertaken on medical data. Medical analysis is one of

the important aspects of our life. But it is impossible to give exact definitions and symptoms of medical

concepts and relationship between the concepts in most of the cases. The boundaries are not clear. The

uncertain nature of medical field needs the use of Fuzzy Logic or its combination with other AI

techniques.

Güney(2016) emphasized in his paper , Fuzzy Logic(FL) is a method of computing that follows human

cognitive ability. FL acts like the way of decision making in humans which based on degrees of truth

instead of Boolean logic YES and NO. The idea of fuzzy logic was introduced by Dr.LotfiZadeh of the

University of California in the 1960s when he was working on the problem of computer understanding of

natural language. He highlighted that human decision making includes a lot of possibilities between YES

and NO, such as possibly, certainly, almost.The fuzzy logic takes possibilities of input to achieve the

distinct output.

Consider the statement, ―If the back pains severe and the patient isold, then apply acupuncture to a

certain point for a long time .‖ To process and model this statement in a computer system, we need more

than programming skills and true‑false statements. All the terms we need to model that is severe, old,

certain point and long‑time are vague and fuzzy. For example, old as a variable is expressed such as the

age group above 50 years, 60 years or 70 years. In the same way here for time variable we may go for

different time chunks. For this reason, medical data analysis applications need to employ methodologies

with fuzzy logic.

Categories of Medical Data

Medical data can be divided based on its attributes. In Fig-1, shows the different data we have considered

for this paper. We may have following types of medical data or it can be combinations of them:

Textual or narrative medical data such as social or family history, answers or description

provided by the patient

Numerical Measurements such as lab results, and other observations

Recorded signals such as ECG, graphical tracing

Pictures of CT scan images, radiologic images

International Journal of Pure and Applied Mathematics Special Issue

16322

Heterogeneous Medical Data, combinations of different types of data

Types of Medical Records

Based on the source of data and who maintains the data we have different types of medical records.

Electronic medical records (EMRs) are digital versions of the notes and information collected by and for

the clinicians in that office, clinic, or hospital.Electronic health records (EHRs) contain information

from all the clinicians involved in a patient’s care and all authorized clinicians involved in a patient's care

can access the information to provide care to that patient. EHRs also share information with other

healthcare providers, such as laboratories and specialists. EHRs follow patients – to the specialist, the

hospital, the nursing home, or even across the country.Personal health records (PHRs)contain the same

types of information as EHRs—diagnoses, medications, immunizations, family medical histories, and

provider contact information—but are designed to be set up, accessed, and managed by patients. PHRs

can include information from a variety of sources including clinicians, home monitoring devices, and

patients themselves.

Figure 1.Categories of Medical Data

Fuzzy Inference System

There are mainly four functional blocks as shown in Fig-2. It contains a knowledge base which consists of

rule base and database. Rule Base contains fuzzy if-then rules and database defines the membership

functions of fuzzy sets used in fuzzy rules.Fuzzification interface unit converts the crisp quantities into

fuzzy quantities.De-Fuzzification interface unit converts the fuzzy quantities into crisp quantities.

To construct a FIS first, we have to define linguistic variables and terms. Then for each linguistic

variables, we have to define membership function. The next task is the construction of membership

functions and then by using these membership functions construction of fuzzy rules. Predefined facts

about the world and expert's knowledge are saved in the database. Then fuzzification takes place means

crisp input will be converted into fuzzy data sets using membership functions. Then fuzzy inference

Textual Data Experimental Data

Imaging Data Signal Data

Medical Data


16323

system which performs decision making task starts. It evaluates the rules in the rule base and combines

results from each rule. The last step is de-fuzzification where output data is converted into non-fuzzy

values.

There are two common fuzzy inference methods used in medical diagnosis systems. The first one is

Mamdani's fuzzy inference method proposed in 1975 by EbrahimMamdani.The second method is Takagi-

Sugeno-Kang, introduced in 1985. Blej and Mostafa (2016) highlighted the differences between these two

FISs. Sugeno inference method is modified version of Mamdani but both of them have their own pros and

cons. Mamdani fuzzy inference system entails a substantial computational burden whereas Sugeno

inference method is computationally efficient. Mamdani is well suited to human input whereas the latter

is well suited to mathematically analysis. The main difference lies in the output. Mamdani gives an output

that is a fuzzy set whereas Sugeno gives an output that is either constant or a linear mathematical

expression. A fuzzy rule in Mamdani can be defined as "If A is X1, and B is X2, then C is X3"where,

X1, X2, X3 are fuzzy sets. But in Sugeno, it is defined as" If A is X1 and B is X2 then C = ax1 + bx2 + c

―where a, b and c are constants.

Figure 2.Functional Blocks of FIS

Different types of fuzzy membership functions

Crisp Input Fuzzification

Decision

Making Data Base

Fuzzy Rules

De-Fuzzification

Crisp Output


16324

We know the first stage of inference system is to select suitable membership functions. There are

different MFs such as crisp MFs, type-1 (T1) fuzzy MFs, and type-2 (T2) fuzzy MFs. A T1 fuzzy MF is

considered as a continuous function on the constituent features. A fuzzy relation of higher type (e.g., type-

2) has been regarded as one way to increase the fuzziness of a relation. Based on data distribution

different application we can use different MFs. Raj, Gupta, Garg,Tanna , Rhee , and Frank (2017)

explained in their work ,when the data follows a uniform random distribution, a crisp MF best represents

the data set. For a crisp MF, every element in the fuzzy set has an equal probability of occurrence, hence

satisfying the requirement of a uniform random distribution. When the data distribution closely represents

a continuous function of the constituent features without many deviations, a T1 fuzzy MF may be best

suited for its representation.Further, if flexibility is required with each instance of the generated data, an

IT2 fuzzy MF may be used as it incorporates a uniform uncertainty with the primary membership. When

the data distribution is such that it vaguely follows a well-defined function on the constituent features and

a significant number of data samples deviate from it. Here it may be useful to represent the data using a

T2 fuzzyMembership Function.

The next section describes the application of FL on different types of medical data.Here we consider the

broad categories based on attributes of input.

FUZZY LOGIC TASKS IN MEDICAL DATA

Singh (2013) explained fuzzy logic is excessively helpful for people involved in research and

development including engineers, mathematicians, computer software developers and researchers, natural

scientists, medical researchers, social scientists, public policy analysts, business analysts, and jurists.

Fuzzy logic has been used in numerous applications such as facial pattern recognition, air conditioners,

washing machines, vacuum cleaners, antiskid braking systems, transmission systems, control of subway

systems and unmanned helicopters, knowledge-based systems for multi-objective optimization of power

systems, weather forecasting systems, models for new product pricing or project risk assessment, medical

diagnosis and treatment plans, and stock trading. Fuzzy logic has been successfully used in numerous

fields such as control systems engineering, image processing, power engineering, industrial automation,

robotics, consumer electronics, and optimization. From NCBI (National Center for Biotechnology

Information) we identified the number of publications in healthcare using fuzzy logic increases every

year. Here Figure3. shows the number of publications from the year 2010 to 2016.Here we have used the

keyword ―fuzzy logic‖ to get the count of publications.

Torres andNieto (2006) explained in their paper diagnosis of the disease involves several levels of

uncertainty and inaccuracy. A single disease may appear in many forms based on the patient, and with

different intensities. A single symptom may correspond to different diseases. The description of disease


16325

entities uses linguistic terms that are also not exact and vague. To deal with inaccuracy and uncertainty,

fuzzy logic introduces fractional truth values, between YES and NO.

In many real-life applications, it is convenient to consider fractional logical values. Consider the

statement "I am healthy". Is it true if I have only one cavity tooth? Is it false if I have a chronic disease?

Everybody is healthy to some degree (say h) and unhealthy to some degree (say u). If we are totally

healthy, then of the value of h = 1, u = 0. Usually, everybody has some minor health problems and h< 1,

but h + u=1.In the other extreme situation, h = 0, and u = 1 so that we are unhealthy i.e. we are dead. In

the case we have only a cavity tooth, we may write h = 0.999,u = 0.001; if we have a painful gastric ulcer,

u = 0.6,h = 0.4, but in the case we have a cancer, probably u = 0.95, h = 0.05.FL has a wide area of tasks

in health care. In Figure-4, listed out few tasks, which take help of fuzzy logic.

Ranking Analysis

In healthcare fuzzy logic is also used for ranking studies. Ranking studies are nothing but ordering

alternatives from best to worst. In healthcare, alternatives mean different types of tests, risk factors, the

performance of health cares, different attributes related to a particular disease, which are crucial for

decision making. T. Kempowsky‑Hamonet al. (2015) applied the fuzzy logic selection on breast cancer

databases and obtained four new gene signatures. D Tadic , M Stefanovic and A Aleksic (2014) proposed

a fuzzy multi-criteria decision-making approach to evaluate suppliers of one kind of medical device with

respect to numerous criteria . Duncan Rangela,Pamplona Salomon (2015) proved that TODIM is better

for ranking analysis than Fuzzy Set Theory(FST) . M Shrief, W Al‑Atabany and M El‑Wakad(2015)

given a new quantitative ranking model based on multi-criteria decision making using fuzzy logic to rank

the computed tomography departments in hospitals. The system is based on factors extracted from both

the hospitals and the CT scan devices.


16326

Figure 3. Publications Using Fuzzy Logic from 2010-2016

Clustering Analysis

Clustering is a process of partitioning a set of data into a set of meaningful sub-classes known as clusters,

in which objects within the same cluster have similar properties and objects of different clusters have

different properties. Clustering cancers cells, genes, and images are the main areas of clustering in health

care. The studies of DPriya, Krithiga , Pavithra and Rajesh Kumar (2015) states that by using modified

fuzzy C-mean clustering algorithm leukaemia in blood microscopic images can be detected. Sharma and

Wasson (2015) published a paper which highlighted a method based on fuzzy rules to segment retinal

blood vessels. This proposed method makes use of the different set of fuzzy rules to process retinal

images taken from publically available DRIVE data set. Tran andNguyen(2016)proposed a unified

framework using Clustering and Fuzzy Rule-based systems for the diagnosis of dental problems, which

shows improvements on the side of classification and decision making.

Prediction Analysis

Predictive analytics is the branch of the advanced analytics which is used to make predictions about

unknown future events. Predictive analytics uses many techniques from data mining, statistics, modelling,

machine learning, and artificial intelligence to analyze current data to make predictions about future.From

1115

1156

1207

1275

1426

1282

1058

0 500 1000 1500

2010

2011

2012

2013

2014

2015

2016

No. of publications

YE

AR


16327

predicting medical issues before they start to provide better treatment programs for patients, predictive

analytics is poised to revolutionize the healthcare industry.

FarzanaIslamet al.(2017) used Adaptive Neuro-fuzzy inference system with a fuzzy C-mean classifier to

predict risk for stroke which is helpful for medical experts.Nilashia ,Ibrahima,Ahmadic and Shahmoradic

(2017) proposed a knowledge-based system using EM, PCA, CART and fuzzy rule-based methods for

classifications using fuzzy logic to improve the prediction accuracy of breast cancer . Vanessa ,Cátia and

Susana (2016) addressedthe prediction of short and long-term mortality in patients that presented Acute

Kidney Injury (AKI) diagnosis at their hospital admission. Fuzzy models are developed using fuzzy c-

means and Gustafson-Kessel algorithms to predict mortality in the ICU within 24 hours

.Alhaddad,,Mohammed ,Kamel and Hagras(2015) presented an interval type-2 fuzzy logic-based model

which can deal with uncertainties to produce better prediction accuracies by generating rules with one

antecedent to find the effective time instances within the effective sensors in relation to given P300 event.

Classifications

In classification, we try to find group memberships for the known and predefined labels (classes). But in

medical field classification, is the process of transforming descriptions of medical diagnoses and

procedures into universal medical code numbers. Diagnosis codes track diseases and other health

conditions. These codes helps in statistical analysis of diseases and therapeutic actions, reimbursement

(insurance claim) and also in knowledge-based and decision support systems.

Sridhar ,Reddy and Prasad(2015) proposed methodologies , the binding of fuzzy logic with

morphological operator to classify mammograms into more intensity colour parts for the suspicious area

.Cátia ,Salgadoaet al.(2016) proposed an ensemble fuzzy modelling approach to a classification problem

based on subgroups of patients identified by individual characteristics by using fuzzy c-means clustering

algorithm.

Pattern Recognition and Feature Extraction

Pattern recognition focuses on the recognition of patterns and regularities in data. Time series analysis

tries to find patterns and rule depending on time, recognition of medical images belong to this type of

studies.Whereas feature selection is the methodology that finds and eliminates the irrelevant samples in

the given space of the samples to help the decider in the decision‑making process. This method is used

especially to eliminate the unhealthy cells/images/tissues to spot the illness in the patients.

The studies of Banerjee, Keller, Popescuand Skubic(2015) outlined a method to study activities of daily

living of elderly people by constructing fuzzy inference system.Srivastava , Chawla and Singh (2015)


16328

implemented an efficient algorithm for gait based human identification. They considered the features such

as height, hip, neck, the knee of the human silhouette and a model based feature such as area under

Hermite curve of hip and knee. The subject recognition has done using Fuzzy Logic.Rubio,Oscar, and

Sepulveda(2016) implemented a method for detection of early-stage breast cancer.They have collected

mammography images from the mini- MIAS database and passed the gradient operator of images as

Fuzzy variables to recognize area with high tone variation.Herman ,Prasad and Thomas(2017)has

examined the applicability of the T2FL approach to the problem of EEG pattern recognition

.SriparnaSahaet al.(2016) demonstrated an interesting approach to gesture recognition for elderly people

on the basis of gesture analysis and generate alarms, thereby finding significance in elderly healthcare.

They have used the concept of interval type-2 fuzzy logic based classification.Wu CH and Wang

(2015).constructed a cloud-based fuzzy expert system for the risk assessment of chronic kidney disease

(CKD).

Figure 4. Fuzzy Logic Tasks on Medial Data

Figure 5 .shows the flowchart of fuzzy inferences on medical data. The input can be any medical data and

output will be ranking, classification, risk analysis or prediction of disease.

DIAGNOSIS THROUGH MEDICAL IMAGING USING FUZZY LOGIC

Peter (2006) highlighted the importance of fuzzy in medical imaging. According to him medical imaging

is a strong supporting element in medical decision making. Two dimensional or three‑dimensional

medical images are generated by magnetic resonance imaging, computed tomography, digital

mammography, positron emission tomography tests. These images can only be usable in medical decision

Fuzzy Tasks

Clustering

Ranking Analysis

Patteren Recognition

Classif ication

Feature Extraction

Prediction


16329

support by postprocessing.FL techniques are also well used in this area. The medical domain is full of

imprecise conditions and vagueness.There exist noise and inaccuracies in edge detection, so we can use

fuzzy logic with traditional methods to detect edges efficiently. Pattern recognition is the way of

determining the patterns, cells, images, etc. Classification and clustering are common techniques used for

pattern recognition. The major fuzzy methods in medical image pattern recognition are fuzzy clustering,

fuzzy rule‑based methods, fuzzy pattern‑matching methods, and methods based on fuzzy relations. In

fuzzy clustering, the fuzzy c‑mean algorithm is one of the most used techniques. Some examples of

FL‑supported medical imaging applications are supported in the diagnosis of brain tumor, classification

of radiographic images, edges detection, images thresholding, motion detection, ranking segmentation

paths.

Figure 6.shows the general framework of fuzzy on images. This describes the input will be features of an

image or grayscale image or Histogram of that input image. After inferencing, the output will be any

decisive measure either classes or segments.

According to Wang, Li. , Qin and Hao (2015) modalities of medical images convey different

information about the human body, organs, and cells, and have their own uses. For example, computed

tomography (CT) images can depict dense structures like bones and hard tissue with less distortion, while

magnetic resonance imaging (MRI) images are better visualized in the case of soft tissues.Whereas T1-

MRI images provide anatomical structure details of tissues, while T2-MRI images provide information

about normal and pathological tissues.

Mohajerani and Ntziachristos(2016) proposed a method to improve imaging performance of fluorescence

molecular tomography (FMT).They proposed a new approach for utilizing prior information, using a

weighted least square (WLS) approach, where the weights were optimized using a Mamdani-type fuzzy

inference system.


16330

Figure 5. Flowchart for Fuzzy Inferences on Medical Data

Yong Yang et al. (2016)proposed a novel multimodal medical image fusion method that adopts a

multiscale geometric analysis of the non -subsampled contourlet transform (NSCT) with type-2 fuzzy

logic techniques. First, the NSCT was performed on preregistered source images to obtain their high- and

low-frequency sub-bands. Next, an effective type-2 fuzzy logic-based fused rule is proposed for fusion of

the high-frequency sub-bands. In the presented fusion approach, the local type-2 fuzzy entropy is

introduced to automatically select high-frequency coefficients. However, for the low-frequency sub-

bands, they were fused by a local energy algorithm based on the corresponding image's local features.

Finally, the fused image was constructed by the inverse NSCT with all composite sub bands. Both

subjective and objective evaluations showed better contrast, accuracy, and versatility in the proposed

approach compared with state-of-the-art methods. Besides, an effective color medical image fusion

scheme is also given in this paper that can inhibit color distortion to a large extent and produce an

improved visual effect.

(Text, Image, Signal or

Experimental )Medical

Data

Fuzzification

Fuzzy

Inference Data Base

Fuzzy Rules

De-Fuzzification

Ranking, Classification, Risk

analysis or Prediction of Disease


16331

FUZZY IN DIAGNOSIS THOUGH SIGNAL PROCESSING

To provide more comfortable and effective healthcare services, a recent trend of healthcare has

been directed towards deinstitutionalization, community care, and home care. The quality of community

and home health care has been significantly improved and many portable devices have also been

developed for a wide variety of applications where signal processing-based software plays a pivotal role

in their success.

Zalabarria,Irigoyen,Martíneaquel and Ramirez (2016) proposed a system for stress detection and

monetarization. This work uses physiological variables such as the electrocardiogram (ECG), the galvanic

skin response (GSR) and the respiration (RSP) in order to estimate the level and classify the type of

stress. On that purpose, an algorithm based on fuzzy logic has been implemented. This computer-

intelligent technique has been combined with a structured processing shaped in the state machine. This

processing classifies stress into 3 different phases or states: alarm, continued stress and relax .

Plerou,Vlamou, and Papadopoulos (2016)concerned with the evaluation of EEG signal analysis

using several pattern recognition methods, and compared analysis to linear and nonlinear pattern

recognition for enhanced efficiency of fuzzy logic systems has been carried out.

Yang , Deng , Choi ,Wang and Takagi–Sugeno–Kang(2016) proposed an important approach to

the detection of epilepsy. They proposed to construct a Takagi-Sugeno-Kang (TSK) FLS based on

transductive transfer learning for identifying epileptic EEG signals. The main objective is to increase the

performance of the epileptic EEG datasets to deal with situations where the training and test datasets

differ with regard to data distribution.

FUZZY LOGIC IN DIAGNOSIS THROUGH TEXTUAL MEDICAL DATA

The majority of medical documents and electronic health records (EHRs) are in text format that

poses a challenge for data processing and finding relevant documents. Looking for ways to automatically

retrieve the enormous amount of health and medical knowledge has always been an intriguing topic.

Powerful methods have been developed in recent years to make the text processing automatic.

Karami,Gangopadhyay, Zhou, and Kharrazi (2017)described fuzzy latent semantic analysis

(FLSA), a novel approach in topic modelling using fuzzy perspective. FLSA can handle health

&medicalcorpora redundancy issue and provides a new method to estimate the number of topics. The

quantitative evaluations show that FLSA produces superior performance and features to Latent Dirichlet

allocation (LDA), the most popular topic model.


16332

Figure 6. Framework of Fuzzy Image Processing

Wang,Zhang , and Xu (2016) proposed a new sentiment computation approach, which is defined

as public sentiments discriminator (PSD), considering both fuzzy logic and sentiment complexity. Unlike

traditional machine learning methods, PSD is based on the rational hypothesis that sentiments are

correlated with each other. A three-level computing structure, sentiment-term level, microblog level and

public sentiment level, is employed. Experiments show that the proposed approach, PSD, can achieve

similar accuracy and FF1-measure but more cognitive results when compared with traditional well-known

machine learning methods. These experimental studies have confirmed that PSD can generate an

interpretable result with no restriction among sentiments.

Karamia et al.(2018) analyzed unstructured health-related text data exchanged via Twitter to

characterize health opinions regarding four common health issues, including diabetes, diet, exercise, and

obesity on a population level. To discover topics from the collected tweets, they used a topic modelling

approach that fuzzy clusters the semantically related words such as assigning "diabetes", "cancer", and

"influenza" into a topic that has an overall "disease" theme.

Najafi., Amirkhani,Papageorgiou,Mosavi and Mohammad (2017) proposed an innovative medical

decision support system by using computing with words in fuzzy cognitive maps. In this paper, all

concepts and the weights of connecting links between them are described based on interval type-2

membership functions (IT2 MFs) expressed as a set of words. In this paper, we utilize CWW FCM to

classify celiac disease (CD), a chronicdisorder.

FIS

Features

Graylevels

Histogram

Classes

Features

Segments


16333

FUZZY LOGIC IN DIAGNOSIS THROUGH NUMERICAL/ EXPERIMENTAL MEDICAL

DATA

Experimental data in science are data produced by a measurement, test method, experimental

design or quasi-experimental design. In clinical research, any data produced are the result of a clinical

trial. Experimental data may be qualitative or quantitative, each being appropriate for different

investigations.

Generally speaking, qualitative data are considered more descriptive and can be subjective in

comparison to having a continuous measurement scale that produces numbers. Whereas quantitative data

are gathered in a manner that is normally experimentally repeatable, qualitative information is usually

more closely related to phenomenal meaning and is, therefore, subject to interpretation by individual

observers.

Chavan, Sambare and Joshi. (2016) proposed a method to recommend a diet based on Prakriti of

person and current season and data is collected from different websites where different dieticians

recommended different diet plans for different Prakriti. They have used Type-2 Fuzzy Logic to handle

uncertainty and Ontology is integrated with fuzzy logic to represent food knowledge for the efficient and

accurate diet recommendation.

Thakur,Raw ,Sharma and Mishra(2016) developeda fuzzy based Inference System in order to

analyze the severity of Thalassemia disease using Fuzzy Logic Toolbox in Matlab by developing 26 if-

then rules. For this three fuzzy input variables such as Mean corpuscular hemoglobin (MCH), Mean

Corpuscular Volume (MCV) and hemoglobin (HGB)wereconsidered.To show the sensitivity of

Thalassemia, three output variables such as minor, intermediate and major were analyzed with input and

rules.

PERFORMANCE ILLUSTRATION

Fuzzy logic with other algorithms and methods is an efficient tool for disease diagnosis. By considering

different works we can list down main application areas in medicine, but not limited to, with their

reported performance accuracy in Table 1.


16334

Table 1

Accuracy of diagnosis of different disease using Fuzzy Logic Techniques

Disease Techniques Reported Accuracy

Breast Cancer

Determination

ANFIS for classification 93.7%

Fuzzy Cognitive Maps

98.3%

CART and Fuzzy rule based system 93.2%

Brain Tumor

Detection

FCM with CSO and OBD 98.3%

HSD with Fuzzy-HKSVM 98.6%

Diabetes Fuzzy Classifier 93.82%

Fusion of Fuzzy Logic, ANN and SVM 95.10%

Heart Disease

Prediction

FCM 92%

Crohn’s disease Neuro Fuzzy Classifier 97.67%

Tuberculosis

diagnosis

ANFIS 97%

Lung Cancer

Detection

Fuzzy Local Information Cluster Means 85.9%

Hybrid Neuro Fuzzy System 95.5%

CONCLUSION

Diagnosis of the disease involves several degrees of uncertainty and vagueness. FL is employed in every

critical decision‑making process of healthcare from supply chain to diagnosis, from mining health data to

retrieve information .Fuzzy nature of the medical decision‑making process makes traditional methods

suffer from elasticity. By using FL we can make system more flexible, robust and efficient by taking into

account all possible values including the blurred ones.Designing an FL system or application requires

moreeffort and time. This makes the computation time for thedesired output longer but it provide more

accurate results in the medical field as it deals with obscurity.


16335

REFERENCES

Alhaddad, MJ.Mohammed A, Kamel M, Hagras H(2015).A genetic interval type‑2 fuzzy logic‑based

approach for generating interpretable linguistic models for the brain P300 phenomena recorded via

brain‑computer interface. Soft Computing ,19:1019‑35.

Amin Dicken,Ruhul.Rubby,S A M Fazle.Naz, Sheefta.ArefinKhaled, A M.Azad, Ashraful.

Rahman,Rashedur M .(2016).Risk Assessment of top five Maglinancies among Males and Females

with respect to occupation, Educational Status, and Smoking Habits. Computer and Information

Science (ICIS), IEEE/ACIS 15th International Conference.

Bagchi, Susama.Huong ,Audrey.(2017). Signal Processing Techniques and Computer-Aided Detection

Systems for Diagnosis of Breast Cancer–A Review Paper. Indian Journal of Science and

Technology ,10.3.

Banerjee ,T. Keller, JM.Popescu, M.Skubic, M.(2015). Recognizing complex instrumental activities of

daily living using scene information and fuzzy logic.Computer Vision and Image Understanding.

Barkana, Buket D. Saricicek,Inci. Yildirim,Burak.(2017).Performance analysis of descriptive statistical

features in retinal vessel segmentation via fuzzy logic, ANN, SVM, and classifier fusion.

Knowledge-Based Systems 118 , 165-176.

Blej,Mohammed.MostafaAzizi.(2016).Comparison of Mamdani-Type and Sugeno-Type Fuzzy Inference

Systems for Fuzzy real time Scheduling. International Journal of Applied Engineering Research,

ISSN 0973-4562 , 11( 22), 11071-11075.

Cátia M. Salgadoa, Susana M. Vieiraa, Luís F. Mendonçaa, StanFinkelsteinc, João M.C.

Sousaa.(2016).Ensemble fuzzy models in personalized health care: Application to

vasopressorsadministration.Engineering Applications of Artificial Intelligence ,49 , 141–148.


16336

CH ,Wu.RK, Chiu.SA,Wang.(2015). A cloud‑based fuzzy expert system for the risk assessment of

chronic kidney diseases.International Journal of Business and System Research ,9,315‑33.

Chavan, Shital V.S. S. Sambare.Joshi,Aniruddha. (2016).Diet recommendation based on Prakriti and

season using Fuzzy ontology and Type-2 Fuzzy Logic. Computing Communication Control and

automation (ICCUBEA), International Conference on. IEEE.

D ,Tadic. M,Stefanovic. A, Aleksic .(2014).The evaluation and ranking of medical device suppliers by

using fuzzy topsis methodology. Journal of Intelligent FuzzySystem .

D, KaviPriya. S R ,Krithiga. P ,Pavithra. J ,Rajesh Kumar.(2015). Detection of leukemia in blood

microscopic images using fuzzy logic.Int. J .Engg. Res. &Sci, &Tech. ,240:197-205.

Duncan Rangela,Luís Alberto.Pamplona Salomon,Valério Antonio.(2015). Comparing rankings from

using TODIM and a fuzzy expert system. Information Technology and Quantitative Management .

Gürsel, Güney. (2016).Healthcare, uncertainty, and fuzzy logic. Retrieved from

http://www.digitmedicine.com.

Herman,Pawel Andrzej.Prasad ,Girijesh. McGinnity,Thomas Martin.(2017,Feb).Designing an Interval

Type-2 Fuzzy Logic System for Handling Uncertainty Effects in Brain-Computer Interface

Classification of Motor Imagery Induced EEG Patterns. IEEE Transactions on Fuzzy Systems ,25(

1).

Islam,Farzana.Binta,Sarah .Shoilee, Alam.Shams,Mithi. M. Rahaman,Rashedur.(2017).Potential risk

factors analysis and risk prediction system for stroke using Fuzzy Logic. Artificial Intelligence

Trends in Intelligent Systems,262-272.

Karami,Amir. Gangopadhyay,Aryya . Zhou,Bin.Kharrazi ,Hadi.(2017,May).Fuzzy Approach Topic

Discovery in Health and Medical Corpora .arXiv:1705.00995v2 (stat.ML).


16337

Karamia,Amir.Dahlb,Alicia A.Turner-McGrievyb, Gabrielle.Kharrazic, Hadi. Shaw ,GeorgeJr.(2018).

Characterizing diabetes, diet, exercise, and obesity comments on Twitter .International Journal of

Information Management, 38 , 1–6.

M ,Shrief. W ,Al‑Atabany.M,El‑Wakad.(2015). Ranking and evaluating CT departments by fuzzy

logic.International Journal Computer Applications , 122, 8-15.

Mohajerani,Pouyan.Ntziachristos,Vasilis.(2016,Feb).An Inversion Scheme for Hybrid Fluorescence

Molecular Tomography Using a Fuzzy Inference System. IEEE Transactions on Medical Imaging,

35(2).

Najafi,Azam.Amirkhani,Abdollah.Papageorgiou,Elpiniki I.Mosavi,Mohammad R.(2017). Medical

decision making based on fuzzy cognitive map and a generalization linguistic weighted power

mean for computing with words.IEEE International Conference on Fuzzy Systems.

Nilashia,Mehrbakhsh.Ibrahima,Othman.Ahmadic,Hossein.Shahmoradic,Leila.(2017,July).A knowledge-

based system for breast cancer classification using the fuzzy logic method. Elsevier Telematics and

Informatics , 34(4), 133–144.

Oscar ,Castillo . Mauricio A,Sanchez. Claudia I, Gonzalez. Gabriela E,Martinez .(2017). Review of

Recent Type-2 Fuzzy Image Processing Applications , www.mdpi.com/journal/information MDPI,

doi:10.3390/info8030097.

Plerou,Antonia.Vlamou,Elena. Papadopoulos, Vasil. (2016,Oct). EEG signal pattern recognition analysis:

fuzzy logic systems ascendancy. Advances in fuzzy sets and systems , 21(2), 107-119 .

Q,Wang. S, Li. H, Qin. A,Hao. (2015,Nov).Robust multi-modal medical image fusion via anisotropic

heat diffusion guided low-rank structural analysis. International Conference on Fusion, 26, 103–

121.


16338

Raj,Desh. Gupta, Aditya.Garg,Bhuvnesh.Tanna,Kenil .Rhee ,Frank Chung-Hoon.(2017). Analysis of

Data Generated from Multidimensional Type-1 and Type-2 Fuzzy Membership Functions.IEEE

Transactions on Fuzzy Systems.

Rogai,Francesco. Manfredi,Claudia. Bocchi,Leonardo.(2016).Metaheuristics for Specialization of a

segmentation algorithm for ultrasound Images. IEEE Transactions on Evolutionary Computation,

20( 5).

Rubio,Yoshio.Oscar, Montiel. Sepulveda,Roberto. (2016). Microcalcification Detection in Mammograms

Based on Fuzzy Logic and Cellular Automata.Nature –Inspired.Design of Hybrid Intelligent

Systems , 667, 583-602.

S,Thakur.Raw,S N. Sharma,R. Mishra,P.(2016).Detection of type of thalassemia disease in patients: A

fuzzy logic approach. International Journal of Applied Pharmaceutical Sciences and Research ,

1(2),88-95.

Saha,Sriparna. Datta,Shreyasi.Konar,Amit.Banerjee,Bonny.Nagar,Atulya K.(2016). A novel gesture

recognition system based on fuzzy logic for healthcare applications. Fuzzy Systems (FUZZ-IEEE),

IEEE International Conference.

Segundo,Unai. Aldámiz-Echevarría,Luis.(2017). Improvement of newborn screening using a fuzzy

inference system. Expert Systems with Applications, 78, 301-318.

Sharma,Sahil.Wasson,Vikas.(2015,Oct).Retinal Blood Vessel Segmentation Using Fuzzy Logic. Journal

of Network Communications and Emerging Technologies , 4( 3).

Singh,Harpreet. Real-Life Applications of Fuzzy Logic.(2013).Advances in Fuzzy Systems, Article ID

581879.


16339

Sridhar, B.Reddy ,KV.Prasad ,AM,.(2015).Mammographic image analysis based on adaptive

morphological fuzzy logic CAD system. International Journal of

BiomedicalEngineeringandTechnology , 17:341‑55.

Srivastava,Smriti.Chawla,Abhishek. Singh,Subhakaran.(2015,Dec). Human Gait Recognition Using

Fuzzy Logic. Computational Intelligence, Cyber Security and Computational Model, 277-287.

Szolovits,Peter .[2006].Medical Informatics: Computer Applications in Health Care . Retrieved from

https://groups.csail.mit.edu / medg /courses / 6872 / 96 / notes / Tsien / sld016.htm .

T,Kempowsky‑Hamon .C,Valle.M,Lacroix‑Triki. L, Hedjazi. L,Trouilh.S,Lamarre.(2015). Fuzzy logic

selection as a new reliable tool to identify molecular grade signatures in breast cancer. The

INNODIAG study.BMC Med Genomics ,8(3).

Torres, Angela. Nieto,Juan J.(2006). Fuzzy Logic in Healthcare and Bioinformatics. Hindawi Publishing

Corporation Journal of Bio health care and Biotechnology , Article ID 91908, 1–7.

Tran ,Manh Tuan.Nguyen ,Hai Minh.(2016,NoV).Medical diagnosis from dental X-ray images: A novel

approach using Clustering combined with Fuzzy Rule-based systems. Annual Conference of the

North American Fuzzy Information Processing Society .

Vanessa ,S. Cunha.Cátia M, Salgado.Susana M. Vieira.(2016) Fuzzy modeling to predict short and long-

term mortality among patients with Acute Kidney Injury. Fuzzy Systems (FUZZ-IEEE)

International Conference.

Wang, Xinzhi.Zhang ,Hui. Xu ,Zheng.(2016,Dec).Public Sentiments Analysis Based on Fuzzy Logic for

Text .International Journal of Software Engineering and Knowledge Engineering, 26(9,10).


16340

Yang ,Changjiang.Deng ,Zhaohong.Choi ,Kup-Sze.Wang, Shitong. Takagi–Sugeno–Kang.(2016,Oct).

Transfer Learning Fuzzy Logic System for the Adaptive Recognition of Epileptic

Electroencephalogram Signals. IEEE Transactions on Fuzzy Systems , 24( 5).

Yang ,Yong.Que,Yue. Huang,Shuying. Lin,Pan.(2016). Multimodal Sensor Medical Image Fusion

Based on Type-2 Fuzzy Logic in NSCT Domain.IEEE Sensors Journal ,16(10).

Zalabarria,Unai.Irigoyen ,Eloy .Martínez,Raquel.Ramirez,Asier Salazar.(2016) .Detection of Stress Level

and Phases by Advanced Physiological Signal Processing Based on Fuzzy Logic. Computational

Intelligence in Security for Information Systems ConferenceICEUTE ,301-312.

Karthigaipriya.T .Srividya.D ROAD ASSET MANAGEMENT SYSTEM FOR MADURAI CBD

International Journal of Innovations in Scientific and Engineering Research (IJISER)


16341

STUDY OF FUZZY LOGIC IN MEDICAL DATA ANALYTICS … · certain point and long time are vague and...

Documents

Transcript of STUDY OF FUZZY LOGIC IN MEDICAL DATA ANALYTICS … · certain point and long time are vague and...